Having read the entire mailing list thread so far, the way this will be resolved is inevitable:
- Ubuntu is going to package Wayland in its main repository and support it. There's no way around it. In fact, see the quote below.
- Either the flavors will use Wayland as the system compositor and completely eschew Mir, or a Mir backend will be added to the version of Weston distributed by Ubuntu and the flavors will use a Wayland compositor running on top of Mir.
There's also some talk in the thread about working to get Mir support integrated into upstream KDE, but all of us who actually follow this situation know that's not going to happen as long as Mir is distro-specific, and there's no evidence that it will ever be anything else. The people suggesting that also doesn't address the problem that Kubuntu is only one flavor, and the Mir problem affects all of the flavors.
Also, there's this nice quote from Canonical employee Marc Deslauriers:
In any case, Kubuntu is quite unlikely to disappear or use a different distro as a base.Wayland will just work in Ubuntu.
*buntu-based distributions just have to use X or Wayland. That is no rule that they must use Mir.
The only advantage of basing on Ubuntu is to gain access to up-to-date packages like Mesa and X, both of which are still at least 1 entire version older even in Debian Sid and Debian Experimental, while still retaining support for apt -get.
If it seems to break a lot of things on the community, the community will not be happy, be it FOSS or the worst closed source software that ever existed, with the exception they're used to closed software caring only about themselves. The same people is sometimes unhappy when FOSS drivers don't meet their requirements and they try to install the binary ones.
HOWEVER, since Mir actually doesn't exclude the possible use of the other environments, that part is a non-problem. The real problem will (likely, but not for sure) for the blob users, since Mir getting drivers will probably imply that Wayland doesn't get them, because it might not be deemed profitable for hardware manufacturers to do so. That and the fragmentation, which is arguable, with too many different point of view, are the real problems Mir introduce.
There's also the fact Mir started with the wrong foot spreading FUD without actually understanding what they were talking about. This not only makes a snowball of retaliation FUD, but makes them lose credibility with the community.
I personnally use Kubuntu, and I really like it.
For me the ubuntu part of Kubuntu is to have KDE software with up to date packages like Ubuntu (and common start system, ...)
I don't switch to other distro because Ubuntu has a good documentation, is compatible with every programs I want to use (Steam, ...), has up to date packages (but not too unstable) and because after using it some years, I've learned many things and useful commands that I fear having to learn again if I change distro.
I've looked at the documentation of Mir and Wayland, and I think Wayland will add more value to my desktop than Mir.
But some others will think that Mir will be better for them (of course when both are ready, we'll compare the scenarios to see which is better in some using case).
Since Keep X - Choose Mir - Choose Wayland are non perfect solutions to please the Kubuntu users, why not split into KubuntuMir and KubuntuWayland? Since Mir is not ready and than Kde should soon work in wayland, KubuntuWayland would be ready before KubuntuMir, but once Mir is ready, KubuntuMir would come to life.
I'd equate sticking by Canonical as being like an abused partner staying in an abusive relationship. They should cut their losses and run.
I remember Ubuntu's PR campaign. It wasn't even released when it was sold like the first debian based 'easy' distro ever.
Instead of being there a long ass line of other distris who did it first. Just without the same amount of advertisement.
Mandrake converted a lot of windows users (and they weren't even debian based).
Suse (not debian based either *gasp*)